Re: [ccp4bb] AW: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] real real-space-refinement

2020-07-30 Thread Robert Nicholls 
To paraphrase from our 2018 Acta Cryst D74(6) paper:

“There has been debate as to whether refinement should be performed in real 
space or reciprocal space; both approaches have advantages and disadvantages. 
In real space, refinement can be performed locally, which has advantages for 
computational speed and parallelisation.
[Update: Refmac now has the option to perform “local refinement”, which means 
focusing on just a localised region in order to speed things up – this option 
is available from within the CCP-EM interface.]
Refinement in reciprocal space allows accounting for errors in Fourier 
coefficients more accurately. And, although errors in neighbouring Fourier 
coefficients are correlated, they are less correlated than between proximal 
points in real space. Furthermore, the degree to which errors are correlated 
will depend on the nature of the underlying data (i.e. on the reconstruction 
methods). There is a common misconception that the original data are in real 
space. Reconstructions are typically performed in Fourier space using the 
projection-slice theorem, before the reconstructed maps are subsequently 
calculated in real space. Alternatively, 3D reconstruction may be performed in 
real space using back-projection, noting that this results in a large 
correlation radius of errors in real space. The procedures in real and 
reciprocal space are mathematically equivalent; in both cases refinement should 
be equivalent (apart from details of implementation). The community would 
benefit from clarification about best practice on this controversial topic, 
thus proper analysis will be required in future."

Indeed, until works have been published on the objective comparative analysis 
of the performance of real vs reciprocal space refinement of models against 
cryo-EM maps, it seems premature to sit too strongly on one side of the fence 
or the other. Until then, I'd recommend people to try different tools and use 
what works best for them, without too much preconception.

Regards,
Rob


Dr Rob Nicholls
Senior Investigator Scientist
MRC Laboratory of Molecular Biology
Francis Crick Avenue
Cambridge Biomedical Campus
Cambridge CB2 0QH



> On 30 Jul 2020, at 13:28, Schreuder, Herman /DE  
> wrote:
> 
> Hi everybody,
> 
> Thank you very much for the many tips! I discovered that in Coot by clicking 
> on the N- and C-terminus of a protein chain, one can indeed run a real-space 
> refinement on the complete chain and the same is true for a rigid body fit. 
> However, subsequent refinement with Refmac did not converge, so it looks like 
> I still have to go manually through the chains. ☹
> Still on my to do list are:
> - test Isolde
> - test Main
> - test the wiggle-fit option in Coot.
> 
> Although the Refmac procedure consisting of converting the cryo-EM map into 
> structure factors and running a "crystallographic" refinement on it seems to 
> work, in my opinion the future will be a robust real-space refinement on the 
> cryo-EM map itself, instead of a refinement on derived structure factors. 
> However, this is my personal opinion.
> 
> Best regards,
> Herman
> 
> -Ursprüngliche Nachricht-
> Von: CCP4 bulletin board  Im Auftrag von Tom Burnley - 
> UKRI STFC
> Gesendet: Mittwoch, 29. Juli 2020 18:51
> An: CCP4BB@JISCMAIL.AC.UK
> Betreff: Re: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] real real-space-refinement
> 
> EXTERNAL : Real sender is  owner-ccp...@jiscmail.ac.uk   
> 
> 
> 
> Hi Herman,
> 
> 
> There is lots of information on CCP-EM's modelling tools here:
> 
> 
> https://urldefense.proofpoint.com/v2/url?u=https-3A__www.ccpem.ac.uk_training_icknield-5F2019_icknield-5F2019.php=DwIFAg=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ=aaxzRy8JT0AM0_MqNvU2fr6IB4wIbkY8SsgvQkYcw_w=D5VOMK2Fz8AOV1PCvJpD9tm914KUxJlsRNRWXYPb7tI=
> 
> 
> This includes a tutorial on Refmac.  As you note this is done in reciprocal 
> space but the CCP-EM pipeline handles conversions from MRC to MTZ format (and 
> runs Refmac via CCP4).
> 
> 
> All the best,
> 
> 
> Tom
> 
> 
> From: CCP4 bulletin board  on behalf of Boaz Shaanan 
> 
> Sent: 29 July 2020 16:31:46
> To: ccp4bb
> Subject: Re: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] real real-space-refinement
> 
> Hi,
> Try searching ccp-em.
> Boaz
> 
> Boaz Shaanan, Ph.D.
> Department of Life Sciences
> Ben Gurion University of the Negev
> Beer Sheva
> Israel
> 
> On Jul 29, 2020 18:29, "Schreuder, Herman /DE"  
> wrote:
> Coot does it, but if one wants to do it for a complete protein, it is a lot 
> of clicking. On the other hand, I have not tried to just select the N- and 
> C-terminus for real-space refinement to see what happens. I would prefer to 
> have some script to do it.
> Best, Herman
> 
> Von: Eleanor Dodson 
> Gesendet: Mittwoch, 29. Juli 2020 17:23
> An: Schreuder, Herman /DE 
> Cc: CCP4BB@JISCMAIL.AC.UK
> Betreff: [EXTERNAL] Re: [ccp4bb] real real-space-refinement
> 
> 
> EXTERNAL :

[ccp4bb] AW: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] real real-space-refinement

2020-07-30 Thread Schreuder, Herman /DE 
Thank you. I think I will then first try to optimize the weight and then 
increase the cycles.
Herman

-Ursprüngliche Nachricht-
Von: Tom Burnley - UKRI STFC  
Gesendet: Donnerstag, 30. Juli 2020 14:35
An: ccp4bb@jiscmail.ac.uk; Schreuder, Herman /DE 
Betreff: Re: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] real real-space-refinement

EXTERNAL : Real sender is  tom.burn...@stfc.ac.uk   



Hi Herman,

Just a couple of tips for Refmac...

1) increase number of cycles - we find some cases require a lot of cycles, more 
than you may be used from MX refinement.
2) optimise weight - this is important and at present the autoweighting option 
is not as reliable for EM refinement as MX.

You could run these in parallel whilst exploring the other very nice tools 
suggested !

Cheers,
Tom





From: CCP4 bulletin board  on behalf of Schreuder, 
Herman /DE 

Sent: 30 July 2020 13:28:00

To: ccp4bb

Subject: [ccp4bb] AW: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] real 
real-space-refinement



Hi everybody,



Thank you very much for the many tips! I discovered that in Coot by clicking on 
the N- and C-terminus of a protein chain, one can indeed run a real-space 
refinement on the complete chain and the same is true for a rigid body fit. 
However, subsequent refinement with Refmac did not converge, so it looks like I 
still have to go manually through the chains. ☹

Still on my to do list are:

- test Isolde

- test Main

- test the wiggle-fit option in Coot.



Although the Refmac procedure consisting of converting the cryo-EM map into 
structure factors and running a "crystallographic" refinement on it seems to 
work, in my opinion the future will be a robust real-space refinement on the 
cryo-EM map itself, instead of a refinement on derived structure factors. 
However, this is my personal opinion.



Best regards,

Herman



-Ursprüngliche Nachricht-

Von: CCP4 bulletin board  Im Auftrag von Tom Burnley - 
UKRI STFC

Gesendet: Mittwoch, 29. Juli 2020 18:51

An: CCP4BB@JISCMAIL.AC.UK

Betreff: Re: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] real real-space-refinement



EXTERNAL : Real sender is  owner-ccp...@jiscmail.ac.uk







Hi Herman,





There is lots of information on CCP-EM's modelling tools here:





https://urldefense.proofpoint.com/v2/url?u=https-3A__www.ccpem.ac.uk_training_icknield-5F2019_icknield-5F2019.php=DwIFAg=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ=aaxzRy8JT0AM0_MqNvU2fr6IB4wIbkY8SsgvQkYcw_w=D5VOMK2Fz8AOV1PCvJpD9tm914KUxJlsRNRWXYPb7tI=





This includes a tutorial on Refmac.  As you note this is done in reciprocal 
space but the CCP-EM pipeline handles conversions from MRC to MTZ format (and 
runs Refmac via CCP4).





All the best,





Tom





From: CCP4 bulletin board  on behalf of Boaz Shaanan 


Sent: 29 July 2020 16:31:46

To: ccp4bb

Subject: Re: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] real real-space-refinement



Hi,

Try searching ccp-em.

Boaz



Boaz Shaanan, Ph.D.

Department of Life Sciences

Ben Gurion University of the Negev

Beer Sheva

Israel



On Jul 29, 2020 18:29, "Schreuder, Herman /DE"  
wrote:

Coot does it, but if one wants to do it for a complete protein, it is a lot of 
clicking. On the other hand, I have not tried to just select the N- and 
C-terminus for real-space refinement to see what happens. I would prefer to 
have some script to do it.

Best, Herman



Von: Eleanor Dodson 

Gesendet: Mittwoch, 29. Juli 2020 17:23

An: Schreuder, Herman /DE 

Cc: CCP4BB@JISCMAIL.AC.UK

Betreff: [EXTERNAL] Re: [ccp4bb] real real-space-refinement





EXTERNAL : Real sender is 
eleanor.dod...@york.ac.uk<mailto:eleanor.dod...@york.ac.uk>



Wont coot do that?

Eleanor



On Wed, 29 Jul 2020 at 16:20, Schreuder, Herman /DE 
mailto:herman.schreu...@sanofi.com>> wrote:

Dear BB,



I would like to do a real real-space-refinement of a protein against a cryo-EM 
map; not the mtz-based Refmac approach. A quick internet search produced a lot 
of Phenix hits, but little ccp4 hits. Does somebody know how to do this using 
ccp4 programs, or has someone a Coot script to do this?



Thank you for your help!

Herman







To unsubscribe from the CCP4BB list, click the following link:

https://urldefense.proofpoint.com/v2/url?u=https-3A__www.jiscmail.ac.uk_cgi-2Dbin_WA-2DJISC.exe-3FSUBED1-3DCCP4BB-26A-3D1=DwIFAg=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ=aaxzRy8JT0AM0_MqNvU2fr6IB4wIbkY8SsgvQkYcw_w=oaC9nUo2Fbh2pGf7qGeQPpoRSR8vFBLjOxlTLJRAxcU=<https://urldefense.proofpoint.com/v2/url?u=https-3A__www.jiscmail.ac.uk_cgi-2Dbin_WA-2DJISC.exe-3FSUBED1-3DCCP4BB-26A-3D1=DwMFaQ=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ=Ch0pjy23vd6uA_krOf8kvGcBAXoSM6A4B64IAhROL4g=12CS6Fb_HI-kbl4I1FebLJB9l_zTdFKUTIGdO53-Ar8=>



_

[ccp4bb] AW: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] real real-space-refinement

2020-07-30 Thread Schreuder, Herman /DE 
Hi everybody,

Thank you very much for the many tips! I discovered that in Coot by clicking on 
the N- and C-terminus of a protein chain, one can indeed run a real-space 
refinement on the complete chain and the same is true for a rigid body fit. 
However, subsequent refinement with Refmac did not converge, so it looks like I 
still have to go manually through the chains. ☹
Still on my to do list are:
- test Isolde
- test Main
- test the wiggle-fit option in Coot.

Although the Refmac procedure consisting of converting the cryo-EM map into 
structure factors and running a "crystallographic" refinement on it seems to 
work, in my opinion the future will be a robust real-space refinement on the 
cryo-EM map itself, instead of a refinement on derived structure factors. 
However, this is my personal opinion.

Best regards,
Herman

-Ursprüngliche Nachricht-
Von: CCP4 bulletin board  Im Auftrag von Tom Burnley - 
UKRI STFC
Gesendet: Mittwoch, 29. Juli 2020 18:51
An: CCP4BB@JISCMAIL.AC.UK
Betreff: Re: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] real real-space-refinement

EXTERNAL : Real sender is  owner-ccp...@jiscmail.ac.uk   



Hi Herman,


There is lots of information on CCP-EM's modelling tools here:


https://urldefense.proofpoint.com/v2/url?u=https-3A__www.ccpem.ac.uk_training_icknield-5F2019_icknield-5F2019.php=DwIFAg=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ=aaxzRy8JT0AM0_MqNvU2fr6IB4wIbkY8SsgvQkYcw_w=D5VOMK2Fz8AOV1PCvJpD9tm914KUxJlsRNRWXYPb7tI=


This includes a tutorial on Refmac.  As you note this is done in reciprocal 
space but the CCP-EM pipeline handles conversions from MRC to MTZ format (and 
runs Refmac via CCP4).


All the best,


Tom


From: CCP4 bulletin board  on behalf of Boaz Shaanan 

Sent: 29 July 2020 16:31:46
To: ccp4bb
Subject: Re: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] real real-space-refinement

Hi,
Try searching ccp-em.
Boaz

Boaz Shaanan, Ph.D.
Department of Life Sciences
Ben Gurion University of the Negev
Beer Sheva
Israel

On Jul 29, 2020 18:29, "Schreuder, Herman /DE"  
wrote:
Coot does it, but if one wants to do it for a complete protein, it is a lot of 
clicking. On the other hand, I have not tried to just select the N- and 
C-terminus for real-space refinement to see what happens. I would prefer to 
have some script to do it.
Best, Herman

Von: Eleanor Dodson 
Gesendet: Mittwoch, 29. Juli 2020 17:23
An: Schreuder, Herman /DE 
Cc: CCP4BB@JISCMAIL.AC.UK
Betreff: [EXTERNAL] Re: [ccp4bb] real real-space-refinement


EXTERNAL : Real sender is 
eleanor.dod...@york.ac.uk

Wont coot do that?
Eleanor

On Wed, 29 Jul 2020 at 16:20, Schreuder, Herman /DE 
mailto:herman.schreu...@sanofi.com>> wrote:
Dear BB,

I would like to do a real real-space-refinement of a protein against a cryo-EM 
map; not the mtz-based Refmac approach. A quick internet search produced a lot 
of Phenix hits, but little ccp4 hits. Does somebody know how to do this using 
ccp4 programs, or has someone a Coot script to do this?

Thank you for your help!
Herman



To unsubscribe from the CCP4BB list, click the following link:
https://urldefense.proofpoint.com/v2/url?u=https-3A__www.jiscmail.ac.uk_cgi-2Dbin_WA-2DJISC.exe-3FSUBED1-3DCCP4BB-26A-3D1=DwIFAg=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ=aaxzRy8JT0AM0_MqNvU2fr6IB4wIbkY8SsgvQkYcw_w=oaC9nUo2Fbh2pGf7qGeQPpoRSR8vFBLjOxlTLJRAxcU=



To unsubscribe from the CCP4BB list, click the following link:
https://urldefense.proofpoint.com/v2/url?u=https-3A__www.jiscmail.ac.uk_cgi-2Dbin_WA-2DJISC.exe-3FSUBED1-3DCCP4BB-26A-3D1=DwIFAg=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ=aaxzRy8JT0AM0_MqNvU2fr6IB4wIbkY8SsgvQkYcw_w=oaC9nUo2Fbh2pGf7qGeQPpoRSR8vFBLjOxlTLJRAxcU=



To unsubscribe from the CCP4BB list, click the following link:
https://urldefense.proofpoint.com/v2/url?u=https-3A__www.jiscmail.ac.uk_cgi-2Dbin_WA-2DJISC.exe-3FSUBED1-3DCCP4BB-26A-3D1=DwIFAg=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ=aaxzRy8JT0AM0_MqNvU2fr6IB4wIbkY8SsgvQkYcw_w=oaC9nUo2Fbh2pGf7qGeQPpoRSR8vFBLjOxlTLJRAxcU=

This email and any attachments are intended solely for the use of the named 
recipients. If you are not the intended recipient you must not use, disclose, 
copy or distribute this email or any of its attachments and should notify the 
sender immediately and delete this email from your system. UK Research and 
Innovation (UKRI) has taken every reasonable